A Novel Data‐Driven Framework for Multiscale Design of Ceria‐Based Soot Oxidation Catalysts
Emil Obeid, Bachar Mourched, Khaled Younes, Anne Cordier, Marlu Cesar Steil, Philippe VernouxABSTRACT
Ceria‐based catalysts were systematically investigated using multiscale characterization techniques on both powders and pellets, together with catalytic performance evaluation. A novel data‐driven framework based on polynomial regression models using two reduced principal component metrics, the surface‐electronic component model (SECM) and the compositional component model (CCM) to accurately predict T5% and T50%. SECM, governed by BET surface area, d 50 , Zr content, and electrical conductivity at 300°C ( σ 300 ), reached an optimum near zero, minimizing T5% to 338°C. CCM, mainly controlled by Ce content with contributions from d 50 and σ 350 , further reduced T5% to approximately 262°C, emphasizing the critical role of Ce‐driven oxygen mobility. These findings introduce SECM and CCM tuning as a novel strategy for enhancing soot oxidation performance. Thermomechanical simulations of a full‐scale ceria‐based diesel particulate filter confirmed its structural stability.